MultiCam Transformation Handling[FEATURE] #1095

Some MultiCam datasets could include transformations which need to be implemented.

Tasks:

• Loading of Transformations into the dataset at creation time (types, format, UI for multicam data import)
  • Check with VIAME to figure out a list of supported formats and types of transformations
• Transformation file passing through for pipelines that utilize it
• Transformation handling for overlapping camera regions. I.E. Creation/Editing a detection in one camera should automatically create/edit one in another camera.
  • Determine the current implementation of transformation handling in VIAME
  • See if there is a way to implement this client side instead of relying on Kwiver. Should be easier using linear transformations